同一张表中的当年销售额和上一年销售额？答案

【问题标题】：Current Year Sales and Previous Year Sales in the same table?同一张表中的当年销售额和上一年销售额？
【发布时间】：2020-02-06 02:27:16
【问题描述】：

早安，

我希望将同一周的“上一年销售额”添加到 Teradata 拉取中，但我在编码方面失败了。我为获得该周/年的销售额而编写的基线是：

SELECT a."WEEK_NBR" AS "YearWeek"
, SUM(a."Sales") AS "CurrentYearSales"

FROM "SALESTABLE" AS a

这会导致：

YearWeek  CurrentYearSales

201901    $7,499

201902    $2,300

201903    $6,360


...

202001    $4,500

202002    $9,000

202003    $8,500

如果数据表中有前一年，我希望能够在同一行显示前一年的销售额。这样完成后的表格将如下所示：

YearWeek  CurrentYearSales PriorYearSales

201901    $7,499           NULL

201902    $2,300           NULL

201903    $6,360           NULL
...

202001    $4,500           $7,499

202002    $9,000           $2,300

202003    $8,500           $6,360

当我搜索时，我能找到的只是如何处理本周的数据，但是这对表中的所有记录都可以吗？

【问题讨论】：

标签： datetime teradata

【解决方案1】：

编辑：当您汇总数据时，您需要在加入之前聚合。

假设WEEK_NBR 是数字，您需要一个自联接：

with cte as 
 (
   SELECT a."WEEK_NBR" AS "YearWeek"
    , SUM(a."Sales") AS "CurrentYearSales"
   FROM "SALESTABLE" AS a
   group by 1
 )
select ...
from cte as t1
left join cte as t2
on t2.YearWeek = t1.Ye arWeek- 100

正如@Andrew 所指出的，这也适用于字符串，因为当您将数字添加到字符串或比较数字和字符串时，Teradata 会对浮点数进行自动类型转换（不过我更喜欢编写显式类型转换）

【讨论】：

无论好坏，如果 week_nbr 为 char 或 varchar，Teradata 将在此处进行隐式转换。
抱歉，省略号在哪里？（还编辑了用 t1 替换 a 的查询？）用 cte 作为（ SELECT t1."WEEK_NBR" AS "YearWeek" , SUM(t1."Sales") AS "CurrentYearSales" FROM "SALESTABLE" group by 1 ）选择 SUM(t2 ."Sales") AS "PriorYearSales" from cte as t1 left join cte as t2 on t2.WEEK_NBR = t1.WEEK_NBR - 100
选择类似于select t1.week_nbr as yearweek, t1.CurrentYearSales as CurrentYearSales, t2.CurrentYearSales as PrevYearSales from...。
我收到Executed as Single statement. Failed [5628 : HY000] Column WEEK_NBR not found in t2 or t1. Elapsed time = 00:00:00.568 STATEMENT 1: WITH failed.
您在 CTE 中对其进行了重命名。使用 t1.YearWeek 而不是 t1.Week_Nbr。